comma-lab

Static dashboard generated from scorer-backed artifacts for the Best Write-up track.

Best Track B current_workflow score

3.25

Best Track B current_workflow bytes

1,669,984

Promotions

Rejections

score vs bytes

Track B runs only. Green = timeline promotion events, red = rejected.

promotion ladder

3.62 — 512x384, lanczos/bicubic, crf 23, keyint 32, bframes 4, ref 4
3.56 — 448x336, lanczos/bicubic, crf 23, keyint 32, bframes 4, ref 4
3.56 — 448x336, lanczos/bicubic, crf 23, keyint 48, bframes 4, ref 4
3.54 — 448x336, lanczos/lanczos, crf 23, keyint 48, bframes 4, ref 4
3.33 — 432x324, lanczos/lanczos, crf 23, keyint 48, bframes 4, ref 4
3.25 — 424x318, lanczos/lanczos, crf 23, keyint 48, bframes 4, ref 4

useful negative results

5.73 — robust_current-roi-two-pass-cpu-2026-04-04
4.47 — robust_current-dynamic-main-roi-cpu-2026-04-05
3.44 — robust_current-464x348-cpu-2026-04-04
3.44 — robust_current-cand-416x312-crf23-g48-b4-r4-cpu-2026-04-05
3.43 — robust_current-cand-432x324-crf23-g48-b3-r4-cpu-2026-04-05
3.38 — robust_current-cand-432x324-crf23-g64-b4-r4-cpu-2026-04-05
3.38 — robust_current-cand-426x320-crf23-g48-b4-r4-cpu-2026-04-06
3.32 — robust_current-cand-428x320-crf23-g48-b4-r4-cpu-2026-04-05

promotion accounting

Rule-faithful values below are local estimates from scorer distortions plus honest bytes, not official published scores.

Run	Scale	Current workflow	Current bytes	Rule-faithful estimate	Rule-faithful bytes
robust_current-medium23-cpu-2026-04-03	512x384	3.62	2,819,374	3.618	2,822,418
robust_current-448x336-medium23-cpu-2026-04-03	448x336	3.56	1,978,141	3.563	1,981,185
robust_current-keyint48-cpu-2026-04-04	448x336	3.56	1,901,606	3.562	1,904,650
robust_current-lanczos-lanczos-cpu-2026-04-04	448x336	3.54	1,901,606	3.546	1,904,650
robust_current-432x324-cpu-2026-04-04	432x324	3.33	1,781,129	3.330	1,787,266
robust_current-cand-424x318-crf23-g48-b4-r4-cpu-2026-04-05	424x318	3.25	1,669,984	3.275	1,704,163

working theses

Bitrate placement and resolution remained the strongest honest levers in this scorer region.
BAT00 became useful as a research-only ranking lane, not as a score authority.
ROI-style multi-stream ideas were informative negative results but too expensive in the tested forms.
The 424x318 / keyint48 floor appears locally stable after nearby 3.27 and 3.26 follow-up failures.

key turning points

First honest floor — robust_current-baseline-cpu-2026-04-03 (4.06)
First big win — robust_current-medium23-cpu-2026-04-03 (3.62)
ROI failure — robust_current-dynamic-main-roi-cpu-2026-04-05 (4.47)
Current best floor — robust_current-cand-424x318-crf23-g48-b4-r4-cpu-2026-04-05 (3.25)

timeline

UTC	Type	Summary
2026-04-04T17:46:16.346354+00:00	research	Tiny fixed-ROI two-pass prototype cut bytes but regressed badly on the official scorer, so it stays rejected.
2026-04-04T19:11:30.675723+00:00	promotion	Promoted robust_current 432x324 / medium / 23 / keyint48 / lanczos+lanczos after the tiny local resolution revisit.
2026-04-05T05:53:25.231667+00:00	research	Measured dynamic main-ROI prototype rejected: main ROI was preserved explicitly, but score regressed to 4.47 and bytes rose to about 2.66 MB.
2026-04-05T06:35:49.299885+00:00	verification	Fresh local CPU regression confirmed the restored promoted floor still scores 3.33.
2026-04-05T06:35:49.299885+00:00	research	BAT00 surrogate v2 remained noisy on the full mixed set, but the codec-only subset improved enough to use as a research-only ranking aid.
2026-04-05T07:42:28.847777+00:00	research	BAT00 codec-only surrogate ranked a small codec shortlist, then the top two local CPU candidates were tested sequentially.
2026-04-05T07:42:28.847777+00:00	decision	Local CPU test of surrogate-ranked #1 candidate (432x324 / crf23 / g48 / b3 / r4) scored 3.43 and was rejected.
2026-04-05T07:42:28.847777+00:00	decision	Local CPU test of surrogate-ranked #2 candidate (432x324 / crf23 / g64 / b4 / r4) scored 3.38 and was rejected.
2026-04-05T08:13:15.198319+00:00	promotion	Local CPU test of surrogate-ranked #3 candidate (424x318 / crf23 / g48 / b4 / r4) scored 3.25 and became the new promoted floor.
2026-04-05T08:48:02.864256+00:00	decision	Nearby follow-up on the new 424x318 floor (bframes3) scored 3.27 and was rejected.
2026-04-05T09:47:41.998084+00:00	decision	Nearby follow-up on the 424x318 floor (keyint64) scored 3.26 and was rejected.
2026-04-05T14:29:14.555005+00:00	decision	Lower-resolution BAT00-ranked follow-up at 416x312 scored 3.44 and was rejected.
2026-04-05T15:10:07.425100+00:00	decision	BAT00-ranked nearby-scale follow-up at 428x320 scored 3.32 and was rejected.
2026-04-05T16:07:20.968299+00:00	decision	BAT00-ranked nearby-scale follow-up at 426x320 scored 3.38 and was rejected.